Fellow: Risa Bochner, MD/SUNY Downstate Medical Center/Kings County Hospital Center
Article: Ramgopal S, Janofsky S, Zuckerbraun NS, et al. Risk of Serious Bacterial Infection in Infants Aged ≤ 60 Days Presenting to Emergency Departments with a History of Fever Only. J Pediatr. Jan 2019; 204:191-195.
Summary: This study aimed to determine the association between the presence or absence of fever at time of presentation to the Emergency Department (ED) and serious bacterial infection in infants ≤60 days with fever. The authors performed a secondary analysis of the Pediatric Emergency Care Applied Research Network (PECARN) dataset. They included infants ≤60 days old who had fever (T≥38C) either by history or measured in the ED, were born ≥ 37 weeks of gestation, and did not have any of the following: focal infection, clinical sepsis, antibiotic use in the 4 days prior to presentation, indwelling catheter or shunt, major congenital anomaly, congenital heart disease, chronic lung disease, or conditions affecting the immune system. Serious bacterial infections occurred in 8.8% (108) of infants who were afebrile in the ED and 12.8% (331) of infants who were febrile in the ED. Infants who were afebrile in the ED were less likely than infants who were febrile in the ED to have a serious bacterial infection (RR, 0.68; 95% CI 0.56-0.84). When analyzed by serious bacterial infection type, the risk of UTI was lower in the afebrile in ED group (RR 0.69, 95% CI 0.56-0.84) but not statistically significant for bacteremia or bacterial meningitis. When stratified by age (≤28 days vs >28-60 days), the lower risk of serious bacterial infection in the afebrile in ED group remained significant in both age groups. In infants ≤28 days, the risk of UTI was lower in the afebrile in ED group (RR 0.66, 95% CI, 0.47-0.92). In infants >28 days, the risk of both UTI (RR 0.71, 95% CI, 0.53-0.96) and bacteremia (RR 0.29, 95% CI, 0.10-0.82) were lower in afebrile in ED group.
Key Strengths: By analyzing a dataset collected prospectively from 26 geographically diverse EDs, the authors were able to obtain a large sample size for analysis. The authors used a rigorous method to ensure accurate measurement of their outcomes (documented urine culture, blood culture and one of the following: CSF culture or documentation of telephone contact or medical record review at least 7 days after ED presentation to ensure no subsequent evaluation for bacterial meningitis).
Limitations: The use of antipyretics before ED arrival was not available in the dataset. Receipt of antipyretics prior to ED arrival may make infants more likely to be afebrile at time of presentation. This could potentially bias the results if these infants are misclassified into the afebrile in ED group when they would’ve otherwise been febrile if not for the antipyretics. The method for temperature measurement at home was also not recorded. Inaccurate measurement of temperature may mean that some infants in the afebrile in ED group never actually had a true fever and, consequently, are at lower risk of serious bacterial infection. Including these infants in the analysis may lead to a false association between the exposure and outcome. The decision to draw cultures was at the discretion of the treating provider. There were 2189 infants excluded from the study due to missing cultures. Infants with a history of fever may be less likely to get cultures drawn and therefore more likely to be excluded from the study. This can potentially lead to selection bias. The study does not include a power analysis. There is also no discussion of whether steps were taken to control for potential confounding variables in their analysis.
Major Takeaway: Infants ≤60 days old with a history of fever only are less likely than those with documented fever on ED presentation to have a serious bacterial infection.
Impact on Practice: Since the risk of serious bacterial infection is still substantial (8.8% overall, 7.6% UTI, 1.5% bacteremia, 0.6% bacterial meningitis) in the afebrile in ED group, the study effect size was small, and there is high morbidity associated with untreated serious bacterial infection, the presence or absence of fever at ED presentation should not be used in isolation to determine whether further clinical evaluation is needed.
Article: Ramgopal S, Janofsky S, Zuckerbraun NS, et al. Risk of Serious Bacterial Infection in Infants Aged ≤ 60 Days Presenting to Emergency Departments with a History of Fever Only. J Pediatr. Jan 2019; 204:191-195.
Summary: This study aimed to determine the association between the presence or absence of fever at time of presentation to the Emergency Department (ED) and serious bacterial infection in infants ≤60 days with fever. The authors performed a secondary analysis of the Pediatric Emergency Care Applied Research Network (PECARN) dataset. They included infants ≤60 days old who had fever (T≥38C) either by history or measured in the ED, were born ≥ 37 weeks of gestation, and did not have any of the following: focal infection, clinical sepsis, antibiotic use in the 4 days prior to presentation, indwelling catheter or shunt, major congenital anomaly, congenital heart disease, chronic lung disease, or conditions affecting the immune system. Serious bacterial infections occurred in 8.8% (108) of infants who were afebrile in the ED and 12.8% (331) of infants who were febrile in the ED. Infants who were afebrile in the ED were less likely than infants who were febrile in the ED to have a serious bacterial infection (RR, 0.68; 95% CI 0.56-0.84). When analyzed by serious bacterial infection type, the risk of UTI was lower in the afebrile in ED group (RR 0.69, 95% CI 0.56-0.84) but not statistically significant for bacteremia or bacterial meningitis. When stratified by age (≤28 days vs >28-60 days), the lower risk of serious bacterial infection in the afebrile in ED group remained significant in both age groups. In infants ≤28 days, the risk of UTI was lower in the afebrile in ED group (RR 0.66, 95% CI, 0.47-0.92). In infants >28 days, the risk of both UTI (RR 0.71, 95% CI, 0.53-0.96) and bacteremia (RR 0.29, 95% CI, 0.10-0.82) were lower in afebrile in ED group.
Key Strengths: By analyzing a dataset collected prospectively from 26 geographically diverse EDs, the authors were able to obtain a large sample size for analysis. The authors used a rigorous method to ensure accurate measurement of their outcomes (documented urine culture, blood culture and one of the following: CSF culture or documentation of telephone contact or medical record review at least 7 days after ED presentation to ensure no subsequent evaluation for bacterial meningitis).
Limitations: The use of antipyretics before ED arrival was not available in the dataset. Receipt of antipyretics prior to ED arrival may make infants more likely to be afebrile at time of presentation. This could potentially bias the results if these infants are misclassified into the afebrile in ED group when they would’ve otherwise been febrile if not for the antipyretics. The method for temperature measurement at home was also not recorded. Inaccurate measurement of temperature may mean that some infants in the afebrile in ED group never actually had a true fever and, consequently, are at lower risk of serious bacterial infection. Including these infants in the analysis may lead to a false association between the exposure and outcome. The decision to draw cultures was at the discretion of the treating provider. There were 2189 infants excluded from the study due to missing cultures. Infants with a history of fever may be less likely to get cultures drawn and therefore more likely to be excluded from the study. This can potentially lead to selection bias. The study does not include a power analysis. There is also no discussion of whether steps were taken to control for potential confounding variables in their analysis.
Major Takeaway: Infants ≤60 days old with a history of fever only are less likely than those with documented fever on ED presentation to have a serious bacterial infection.
Impact on Practice: Since the risk of serious bacterial infection is still substantial (8.8% overall, 7.6% UTI, 1.5% bacteremia, 0.6% bacterial meningitis) in the afebrile in ED group, the study effect size was small, and there is high morbidity associated with untreated serious bacterial infection, the presence or absence of fever at ED presentation should not be used in isolation to determine whether further clinical evaluation is needed.